| From: | Douglas Doole <dougdoole(at)gmail(dot)com> |
|---|---|
| To: | Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: WIP: Faster Expression Processing v4 |
| Date: | 2017-03-14 22:31:50 |
| Message-ID: | CADE5jY+=ZDBMcyNzL_HRxy72OdENGerhYt_VyooE1cM05tZBRg@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, Mar 14, 2017 at 3:16 PM Andres Freund <andres(at)anarazel(dot)de> wrote:
> Hm. Right now ExprState's are allocated in several places - but we
> could easily move that to a central place. Have a bit of a hard time
> seing that that branch during *initialization* time is that expensive,
> especially given that previously we allocated a lot of memory separately
> too.
>
I didn't make any comparisons of the cost of the new init against the old
init with this change in particular - I just saw that it made the new init
faster. I also didn't play around to determine if the savings was found in
removing the branch misprediction or inlining or both.
I certainly wouldn't hold up your commit for this, but it's something that
might be worth a second look once the dust has settled.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andres Freund | 2017-03-14 22:47:06 | Re: WIP: Faster Expression Processing v4 |
| Previous Message | Andres Freund | 2017-03-14 22:16:48 | Re: WIP: Faster Expression Processing v4 |